Citation

Branca C, Blasi CD. Fire Safety J. 2008; 43(5): 317-324.

Copyright

(Copyright © 2008, Elsevier Publishing)

DOI

10.1016/j.firesaf.2007.11.004

PMID

unavailable

Abstract

Motivated by the aim of improving the understanding of reaction kinetics, thermogravimetric measurements (heating rates of 5, 10 and 20 K/min up to a temperature of 873 K) in air have been made of wood impregnated with diammonium phosphate and diammonium sulfate and, for comparison purposes, of untreated wood. Wood devolatilization is always followed by char combustion, but the added salts introduce important changes. Indeed, the former is anticipated at lower temperatures with reduced maximum rates, whereas the latter is significantly retarded. For the interpretation of the measured weight loss characteristics, a mechanism has been developed, consisting of parallel reactions with the usual Arrhenius rates and a linear (devolatilization) or a power-law (combustion) dependence on the solid mass fraction. In all cases, three reactions are sufficient for describing the most important features of the devolatilization process, with activation energies practically invariant with the treatments (106, 226 and 90–104 kJ/mol for the decomposition of the pseudo-components hemicellulose, cellulose and lignin, respectively). The effects and differences of the two salts can be effectively taken into account by variations in the pre-exponential factors and amounts of lumped volatile species produced. However, a wide zone of slow volatile release, preceding char combustion, appears, which requires the introduction of two additional reactions with activation energies of 92 and 115 kJ/mol. Keywords: Wood; Flame retardancy; Kinetics; Modeling